Additive manufacturing (AM) is here. This was one of the themes that came out of a September seminar on the subject organised by machine vendor Sweden-headquartered Arcam at the Coventry-based Manufacturing Technology Centre. AM machines are making real production parts in metal (and plastic); 50,000 orthopaedic implants spanning up to 100 types have been made via AM; GE subsidiary Avio Aero and Arcam have developed a way to 3D print turbine blades in gamma titanium aluminide, a lightweight material that withstands the high temperature, high pressure environment of aerospace engines that was previously impossible to form; and Rolls-Royce has made a 1.5 m diameter front bearing housing for the XWB-97 engine, due to be tested this year for the Airbus A350-1000, using AM.
Nigel Bunt, managing director of Arcam Cad to Metal Ltd (01926 491 300), says that, for now, additive manufacturing exists in sweet spots: small market segments particularly favourable to the technology’s unique benefits. Arcam’s EBM technology uses a directed electron beam to melt a thin layer of powder spread across a bed at near-vacuum and high temperature. The process differs from the blown powder method used by DMG Mori and Mazakon metalcutting machine tools that feature AM heads (see ‘Adding plus subtracting,’ Machinery, June 2015, pp. 14-16).
Early days though these might be, AM still has a lot going for it. First, because additive manufacturing technology builds components up from scratch, based on a CAD data, it needs no tooling. Provided the drawing is correct, urgent parts can be cranked out in a matter of hours. This aspect could make AM an interesting competitor to conventional machining processes in an aerospace speedshop environment. Its very speed changes the nature of product design, allowing companies to adopt a more exploratory, iterative design approach in the same time that a traditional process would take to cast a single part in a mould and die operation.
Second, 3D printed parts are near net shape; far closer to the final desired profiles than is a billet (but they are not necessarily perfect; even on the most advanced machines many vendors would still expect to machine AM parts for surface finish).
The high cost of titanium, an increasingly common aerospace material, is becoming a significant cost driver of parts in aerospace. Hogging a part out of a solid titanium block wastes far more material than building it from titanium powder; and titanium swarf is much less valuable than steel or aluminium swarf because exposed surfaces of the metal form the very hard compound titanium dioxide.
Third, because of the way that the additive manufacturing process works, depositing melted powder affects the microstructure of the new component; the process not only creates the part, but also material properties as well. Last year, researchers at the University of Tennessee and Oak Ridge National Laboratory were able to create a material with the letters ‘DOE’ running through the entire solid material thickness, like a stick of Brighton rock, using the Arcam EBM process on a nickel part. “We’re using well established metallurgical phenomena, but we’ve never been able to control the processes well enough to take advantage of them at this scale and at this level of detail,” said Suresh Babu, the University of Tennessee-ORNL governor's chair for advanced manufacturing at the time.
So, in the future, AM designers should be able to tailor the component’s microstructure to better enable it to do its job, something impossible in traditional milling and turning.
“With AM products, people always ask me, how strong is it? The answer is, how strong do you want it to be?” says Rob Sharman, global head of additive manufacturing, GKN Aerospace.
In March, GKN and Arcam entered a strategic partnership to develop and industrialise Arcam’s electron beam melting process. An initial part of that project saw the purchase of two Arcam Q20 machines, the newest model and launched last year. The two will collaborate on EBM equipment “able to manufacture complex titanium structures at the high volumes required to meet future demand.”
Sharman points out that GKN also manufactures powder, through subsidiary GKN Sinter Metals. And given the importance of powder in the process, it is no surprise that it receives particular scrutiny at the new National Centre for Net Shape and Additive Manufacturing, based at the MTC in Coventry (see Machinery August, p20-21), that operates a powder characterisation facility.
The MTC’s goal is to make AM a standard process, moving away from the expert artisanal process it seems to be at the moment. Ross Trepleton, group technology manager, component technology at the MTC says: “We're more about developing the whole manufacturing system around [powder melting in additive manufacturing]. Putting in safe systems of work, operating procedures, health and safety guidance and documents; all these non-sexy process steps, that's what we enjoy doing.”
The final element of AM, and by no means the least technical, is part design. Because the process builds up parts layer by layer, it enables 3D printers to craft wild designs impossible to make any other way, having, for example, complex internal or external features. The ability to produce trabecular – filamentous – structures is perhaps the reason for the success of AM in orthopaedics, since they allow cavities for the bone to grow into.
Sample orthopaedic part with mixed surfaces
AM parts for weight-critical applications such as aerospace or civil engineering could offer similar structural performance to traditionally manufactured parts but much lower weights. An example given at the conference is a bionic-style bridge section devised by Arup with engineering design software company WithinLab, AM partner CRDM/3D Systems and development partner EOS. However, such topologically-optimised designs often face a skeptical reception by internal qualification teams and regulators, who are more concerned with the designs’ inspectability and measurability.
Such designs are also difficult to make, particularly at the moment when multiple different types of software are used to make them, explains Andrew Triantaphyllou, technology area lead, design for AM at the Manufacturing Technology Centre, which is working to simplify the process.
“This workflow involves up to six different software packages, plus the related skill set, cost and time. It is not ideal for the newcomer. What we'd ideally like to see is an integrated environment, just like you see in CAD today; one that has add-ons for injection moulding or sheet metal that tell you how to design. We want something similar for AM parts.”
He concludes: “What we tell people is that AM is good at complexity, but of course it is not as free as they hear in the press. You need to able to make a valid model, you need to know the rules so you can build it, and you need to be able to post-process these complex parts,” – in other words, remove excess powder, machine it for surface quality and inspect it.
Box item
Towards a national AM strategy
Work has begun on developing a UK strategy to accelerate innovation in additive manufacturing.
Says ‘The Case for Additive manufacturing,’ a positioning paper put together and published by some key AM executives in March 2015: “The UK is amongst the global leaders in both the development of knowledge and successful application of AM technology. However, there are gaps in the supply chain (materials supply, equipment, post-processing and validation). Therefore, industry leaders and initiatives such as the 2013 Foresight report [The Future of Manufacturing: A new era of opportunity and challenge for the UK] are recommending the urgent development of a UK National Strategy on AM.”
A series of workshops attended by more than 120 industry representatives earlier this year, plus a website – www.amnationalstrategy.uk – produced about 900 contributions that are now being analysed. Although Machinery understands that a preliminary analysis of the contributions is almost complete, the final strategy document will continue to be developed through next year.
This article was originally published in the November 2015 issue of Machinery magazine.
